mirror of
https://github.com/opencloud-eu/opencloud.git
synced 2025-12-16 17:45:39 -06:00
ed0dbce978
* feat(search): introduce search query package With the increasing complexity of how we organize our resources, the search must also be able to find them using entity properties. The query package provides the necessary functionality to do this. This makes it possible to search for resources via KQL, the microsoft spec is largely covered and can be used for this. In the current state, the legacy query language is still used, in a future update this will be deprecated and KQL will become the standard
1952 lines
45 KiB
Go
1952 lines
45 KiB
Go
// Code generated by pigeon; DO NOT EDIT.
|
|
|
|
package kql
|
|
|
|
import (
|
|
"bytes"
|
|
"errors"
|
|
"fmt"
|
|
"io"
|
|
"math"
|
|
"os"
|
|
"sort"
|
|
"strconv"
|
|
"strings"
|
|
"sync"
|
|
"unicode"
|
|
"unicode/utf8"
|
|
)
|
|
|
|
var g = &grammar{
|
|
rules: []*rule{
|
|
{
|
|
name: "AST",
|
|
pos: position{line: 9, col: 1, offset: 143},
|
|
expr: &actionExpr{
|
|
pos: position{line: 10, col: 5, offset: 154},
|
|
run: (*parser).callonAST1,
|
|
expr: &seqExpr{
|
|
pos: position{line: 10, col: 5, offset: 154},
|
|
exprs: []any{
|
|
&ruleRefExpr{
|
|
pos: position{line: 10, col: 5, offset: 154},
|
|
name: "_",
|
|
},
|
|
&labeledExpr{
|
|
pos: position{line: 10, col: 7, offset: 156},
|
|
label: "nodes",
|
|
expr: &ruleRefExpr{
|
|
pos: position{line: 10, col: 13, offset: 162},
|
|
name: "Nodes",
|
|
},
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 10, col: 19, offset: 168},
|
|
name: "_",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "Nodes",
|
|
pos: position{line: 14, col: 1, offset: 225},
|
|
expr: &actionExpr{
|
|
pos: position{line: 15, col: 5, offset: 238},
|
|
run: (*parser).callonNodes1,
|
|
expr: &labeledExpr{
|
|
pos: position{line: 15, col: 5, offset: 238},
|
|
label: "n",
|
|
expr: &oneOrMoreExpr{
|
|
pos: position{line: 15, col: 7, offset: 240},
|
|
expr: &seqExpr{
|
|
pos: position{line: 16, col: 9, offset: 250},
|
|
exprs: []any{
|
|
&ruleRefExpr{
|
|
pos: position{line: 16, col: 9, offset: 250},
|
|
name: "_",
|
|
},
|
|
&choiceExpr{
|
|
pos: position{line: 18, col: 13, offset: 274},
|
|
alternatives: []any{
|
|
&ruleRefExpr{
|
|
pos: position{line: 18, col: 13, offset: 274},
|
|
name: "GroupNode",
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 19, col: 13, offset: 298},
|
|
name: "PropertyRestrictionNodes",
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 20, col: 13, offset: 337},
|
|
name: "BooleanOperatorNode",
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 21, col: 13, offset: 371},
|
|
name: "FreeTextKeywordNodes",
|
|
},
|
|
},
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 23, col: 9, offset: 410},
|
|
name: "_",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "GroupNode",
|
|
pos: position{line: 32, col: 1, offset: 581},
|
|
expr: &actionExpr{
|
|
pos: position{line: 33, col: 5, offset: 598},
|
|
run: (*parser).callonGroupNode1,
|
|
expr: &seqExpr{
|
|
pos: position{line: 33, col: 5, offset: 598},
|
|
exprs: []any{
|
|
&labeledExpr{
|
|
pos: position{line: 33, col: 5, offset: 598},
|
|
label: "k",
|
|
expr: &zeroOrOneExpr{
|
|
pos: position{line: 33, col: 7, offset: 600},
|
|
expr: &oneOrMoreExpr{
|
|
pos: position{line: 33, col: 8, offset: 601},
|
|
expr: &ruleRefExpr{
|
|
pos: position{line: 33, col: 8, offset: 601},
|
|
name: "Char",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
&zeroOrOneExpr{
|
|
pos: position{line: 33, col: 16, offset: 609},
|
|
expr: &choiceExpr{
|
|
pos: position{line: 33, col: 17, offset: 610},
|
|
alternatives: []any{
|
|
&ruleRefExpr{
|
|
pos: position{line: 33, col: 17, offset: 610},
|
|
name: "ColonOperator",
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 33, col: 33, offset: 626},
|
|
name: "EqualOperator",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
&litMatcher{
|
|
pos: position{line: 33, col: 49, offset: 642},
|
|
val: "(",
|
|
ignoreCase: false,
|
|
want: "\"(\"",
|
|
},
|
|
&labeledExpr{
|
|
pos: position{line: 33, col: 53, offset: 646},
|
|
label: "v",
|
|
expr: &ruleRefExpr{
|
|
pos: position{line: 33, col: 55, offset: 648},
|
|
name: "Nodes",
|
|
},
|
|
},
|
|
&litMatcher{
|
|
pos: position{line: 33, col: 61, offset: 654},
|
|
val: ")",
|
|
ignoreCase: false,
|
|
want: "\")\"",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "PropertyRestrictionNodes",
|
|
pos: position{line: 41, col: 1, offset: 858},
|
|
expr: &choiceExpr{
|
|
pos: position{line: 42, col: 5, offset: 890},
|
|
alternatives: []any{
|
|
&ruleRefExpr{
|
|
pos: position{line: 42, col: 5, offset: 890},
|
|
name: "YesNoPropertyRestrictionNode",
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 43, col: 5, offset: 925},
|
|
name: "TextPropertyRestrictionNode",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "YesNoPropertyRestrictionNode",
|
|
pos: position{line: 45, col: 1, offset: 954},
|
|
expr: &actionExpr{
|
|
pos: position{line: 46, col: 5, offset: 990},
|
|
run: (*parser).callonYesNoPropertyRestrictionNode1,
|
|
expr: &seqExpr{
|
|
pos: position{line: 46, col: 5, offset: 990},
|
|
exprs: []any{
|
|
&labeledExpr{
|
|
pos: position{line: 46, col: 5, offset: 990},
|
|
label: "k",
|
|
expr: &oneOrMoreExpr{
|
|
pos: position{line: 46, col: 7, offset: 992},
|
|
expr: &ruleRefExpr{
|
|
pos: position{line: 46, col: 7, offset: 992},
|
|
name: "Char",
|
|
},
|
|
},
|
|
},
|
|
&choiceExpr{
|
|
pos: position{line: 46, col: 14, offset: 999},
|
|
alternatives: []any{
|
|
&ruleRefExpr{
|
|
pos: position{line: 46, col: 14, offset: 999},
|
|
name: "ColonOperator",
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 46, col: 30, offset: 1015},
|
|
name: "EqualOperator",
|
|
},
|
|
},
|
|
},
|
|
&labeledExpr{
|
|
pos: position{line: 46, col: 45, offset: 1030},
|
|
label: "v",
|
|
expr: &choiceExpr{
|
|
pos: position{line: 46, col: 48, offset: 1033},
|
|
alternatives: []any{
|
|
&litMatcher{
|
|
pos: position{line: 46, col: 48, offset: 1033},
|
|
val: "true",
|
|
ignoreCase: false,
|
|
want: "\"true\"",
|
|
},
|
|
&litMatcher{
|
|
pos: position{line: 46, col: 57, offset: 1042},
|
|
val: "false",
|
|
ignoreCase: false,
|
|
want: "\"false\"",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "TextPropertyRestrictionNode",
|
|
pos: position{line: 50, col: 1, offset: 1112},
|
|
expr: &actionExpr{
|
|
pos: position{line: 51, col: 5, offset: 1147},
|
|
run: (*parser).callonTextPropertyRestrictionNode1,
|
|
expr: &seqExpr{
|
|
pos: position{line: 51, col: 5, offset: 1147},
|
|
exprs: []any{
|
|
&labeledExpr{
|
|
pos: position{line: 51, col: 5, offset: 1147},
|
|
label: "k",
|
|
expr: &oneOrMoreExpr{
|
|
pos: position{line: 51, col: 7, offset: 1149},
|
|
expr: &ruleRefExpr{
|
|
pos: position{line: 51, col: 7, offset: 1149},
|
|
name: "Char",
|
|
},
|
|
},
|
|
},
|
|
&choiceExpr{
|
|
pos: position{line: 51, col: 14, offset: 1156},
|
|
alternatives: []any{
|
|
&ruleRefExpr{
|
|
pos: position{line: 51, col: 14, offset: 1156},
|
|
name: "ColonOperator",
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 51, col: 30, offset: 1172},
|
|
name: "EqualOperator",
|
|
},
|
|
},
|
|
},
|
|
&labeledExpr{
|
|
pos: position{line: 51, col: 45, offset: 1187},
|
|
label: "v",
|
|
expr: &choiceExpr{
|
|
pos: position{line: 51, col: 48, offset: 1190},
|
|
alternatives: []any{
|
|
&ruleRefExpr{
|
|
pos: position{line: 51, col: 48, offset: 1190},
|
|
name: "String",
|
|
},
|
|
&oneOrMoreExpr{
|
|
pos: position{line: 51, col: 57, offset: 1199},
|
|
expr: &charClassMatcher{
|
|
pos: position{line: 51, col: 57, offset: 1199},
|
|
val: "[^ ()]",
|
|
chars: []rune{' ', '(', ')'},
|
|
ignoreCase: false,
|
|
inverted: true,
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "FreeTextKeywordNodes",
|
|
pos: position{line: 60, col: 1, offset: 1406},
|
|
expr: &choiceExpr{
|
|
pos: position{line: 61, col: 5, offset: 1434},
|
|
alternatives: []any{
|
|
&ruleRefExpr{
|
|
pos: position{line: 61, col: 5, offset: 1434},
|
|
name: "PhraseNode",
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 62, col: 5, offset: 1451},
|
|
name: "WordNode",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "PhraseNode",
|
|
pos: position{line: 64, col: 1, offset: 1461},
|
|
expr: &actionExpr{
|
|
pos: position{line: 65, col: 6, offset: 1480},
|
|
run: (*parser).callonPhraseNode1,
|
|
expr: &seqExpr{
|
|
pos: position{line: 65, col: 6, offset: 1480},
|
|
exprs: []any{
|
|
&zeroOrOneExpr{
|
|
pos: position{line: 65, col: 6, offset: 1480},
|
|
expr: &ruleRefExpr{
|
|
pos: position{line: 65, col: 6, offset: 1480},
|
|
name: "ColonOperator",
|
|
},
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 65, col: 21, offset: 1495},
|
|
name: "_",
|
|
},
|
|
&labeledExpr{
|
|
pos: position{line: 65, col: 23, offset: 1497},
|
|
label: "v",
|
|
expr: &ruleRefExpr{
|
|
pos: position{line: 65, col: 25, offset: 1499},
|
|
name: "String",
|
|
},
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 65, col: 32, offset: 1506},
|
|
name: "_",
|
|
},
|
|
&zeroOrOneExpr{
|
|
pos: position{line: 65, col: 34, offset: 1508},
|
|
expr: &ruleRefExpr{
|
|
pos: position{line: 65, col: 34, offset: 1508},
|
|
name: "ColonOperator",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "WordNode",
|
|
pos: position{line: 69, col: 1, offset: 1585},
|
|
expr: &actionExpr{
|
|
pos: position{line: 70, col: 6, offset: 1602},
|
|
run: (*parser).callonWordNode1,
|
|
expr: &seqExpr{
|
|
pos: position{line: 70, col: 6, offset: 1602},
|
|
exprs: []any{
|
|
&zeroOrOneExpr{
|
|
pos: position{line: 70, col: 6, offset: 1602},
|
|
expr: &ruleRefExpr{
|
|
pos: position{line: 70, col: 6, offset: 1602},
|
|
name: "ColonOperator",
|
|
},
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 70, col: 21, offset: 1617},
|
|
name: "_",
|
|
},
|
|
&labeledExpr{
|
|
pos: position{line: 70, col: 23, offset: 1619},
|
|
label: "v",
|
|
expr: &oneOrMoreExpr{
|
|
pos: position{line: 70, col: 25, offset: 1621},
|
|
expr: &charClassMatcher{
|
|
pos: position{line: 70, col: 25, offset: 1621},
|
|
val: "[^ :()]",
|
|
chars: []rune{' ', ':', '(', ')'},
|
|
ignoreCase: false,
|
|
inverted: true,
|
|
},
|
|
},
|
|
},
|
|
&ruleRefExpr{
|
|
pos: position{line: 70, col: 34, offset: 1630},
|
|
name: "_",
|
|
},
|
|
&zeroOrOneExpr{
|
|
pos: position{line: 70, col: 36, offset: 1632},
|
|
expr: &ruleRefExpr{
|
|
pos: position{line: 70, col: 36, offset: 1632},
|
|
name: "ColonOperator",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "BooleanOperatorNode",
|
|
pos: position{line: 78, col: 1, offset: 1837},
|
|
expr: &actionExpr{
|
|
pos: position{line: 79, col: 5, offset: 1864},
|
|
run: (*parser).callonBooleanOperatorNode1,
|
|
expr: &choiceExpr{
|
|
pos: position{line: 79, col: 6, offset: 1865},
|
|
alternatives: []any{
|
|
&litMatcher{
|
|
pos: position{line: 79, col: 6, offset: 1865},
|
|
val: "AND",
|
|
ignoreCase: false,
|
|
want: "\"AND\"",
|
|
},
|
|
&litMatcher{
|
|
pos: position{line: 79, col: 14, offset: 1873},
|
|
val: "OR",
|
|
ignoreCase: false,
|
|
want: "\"OR\"",
|
|
},
|
|
&litMatcher{
|
|
pos: position{line: 79, col: 21, offset: 1880},
|
|
val: "NOT",
|
|
ignoreCase: false,
|
|
want: "\"NOT\"",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "ColonOperator",
|
|
pos: position{line: 83, col: 1, offset: 1944},
|
|
expr: &actionExpr{
|
|
pos: position{line: 84, col: 5, offset: 1965},
|
|
run: (*parser).callonColonOperator1,
|
|
expr: &litMatcher{
|
|
pos: position{line: 84, col: 5, offset: 1965},
|
|
val: ":",
|
|
ignoreCase: false,
|
|
want: "\":\"",
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "EqualOperator",
|
|
pos: position{line: 88, col: 1, offset: 2005},
|
|
expr: &actionExpr{
|
|
pos: position{line: 89, col: 5, offset: 2026},
|
|
run: (*parser).callonEqualOperator1,
|
|
expr: &litMatcher{
|
|
pos: position{line: 89, col: 5, offset: 2026},
|
|
val: "=",
|
|
ignoreCase: false,
|
|
want: "\"=\"",
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "Char",
|
|
pos: position{line: 97, col: 1, offset: 2189},
|
|
expr: &actionExpr{
|
|
pos: position{line: 98, col: 5, offset: 2201},
|
|
run: (*parser).callonChar1,
|
|
expr: &charClassMatcher{
|
|
pos: position{line: 98, col: 5, offset: 2201},
|
|
val: "[A-Za-z]",
|
|
ranges: []rune{'A', 'Z', 'a', 'z'},
|
|
ignoreCase: false,
|
|
inverted: false,
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "String",
|
|
pos: position{line: 102, col: 1, offset: 2246},
|
|
expr: &actionExpr{
|
|
pos: position{line: 103, col: 5, offset: 2260},
|
|
run: (*parser).callonString1,
|
|
expr: &seqExpr{
|
|
pos: position{line: 103, col: 5, offset: 2260},
|
|
exprs: []any{
|
|
&litMatcher{
|
|
pos: position{line: 103, col: 5, offset: 2260},
|
|
val: "\"",
|
|
ignoreCase: false,
|
|
want: "\"\\\"\"",
|
|
},
|
|
&labeledExpr{
|
|
pos: position{line: 103, col: 9, offset: 2264},
|
|
label: "v",
|
|
expr: &zeroOrMoreExpr{
|
|
pos: position{line: 103, col: 11, offset: 2266},
|
|
expr: &charClassMatcher{
|
|
pos: position{line: 103, col: 11, offset: 2266},
|
|
val: "[^\"]",
|
|
chars: []rune{'"'},
|
|
ignoreCase: false,
|
|
inverted: true,
|
|
},
|
|
},
|
|
},
|
|
&litMatcher{
|
|
pos: position{line: 103, col: 17, offset: 2272},
|
|
val: "\"",
|
|
ignoreCase: false,
|
|
want: "\"\\\"\"",
|
|
},
|
|
},
|
|
},
|
|
},
|
|
},
|
|
{
|
|
name: "_",
|
|
pos: position{line: 107, col: 1, offset: 2307},
|
|
expr: &zeroOrMoreExpr{
|
|
pos: position{line: 108, col: 5, offset: 2316},
|
|
expr: &charClassMatcher{
|
|
pos: position{line: 108, col: 5, offset: 2316},
|
|
val: "[ \\t]",
|
|
chars: []rune{' ', '\t'},
|
|
ignoreCase: false,
|
|
inverted: false,
|
|
},
|
|
},
|
|
},
|
|
},
|
|
}
|
|
|
|
func (c *current) onAST1(nodes any) (any, error) {
|
|
return buildAST(nodes, c.text, c.pos)
|
|
|
|
}
|
|
|
|
func (p *parser) callonAST1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onAST1(stack["nodes"])
|
|
}
|
|
|
|
func (c *current) onNodes1(n any) (any, error) {
|
|
return buildNodes(n)
|
|
|
|
}
|
|
|
|
func (p *parser) callonNodes1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onNodes1(stack["n"])
|
|
}
|
|
|
|
func (c *current) onGroupNode1(k, v any) (any, error) {
|
|
return buildGroupNode(k, v, c.text, c.pos)
|
|
|
|
}
|
|
|
|
func (p *parser) callonGroupNode1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onGroupNode1(stack["k"], stack["v"])
|
|
}
|
|
|
|
func (c *current) onYesNoPropertyRestrictionNode1(k, v any) (any, error) {
|
|
return buildBooleanNode(k, v, c.text, c.pos)
|
|
|
|
}
|
|
|
|
func (p *parser) callonYesNoPropertyRestrictionNode1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onYesNoPropertyRestrictionNode1(stack["k"], stack["v"])
|
|
}
|
|
|
|
func (c *current) onTextPropertyRestrictionNode1(k, v any) (any, error) {
|
|
return buildStringNode(k, v, c.text, c.pos)
|
|
|
|
}
|
|
|
|
func (p *parser) callonTextPropertyRestrictionNode1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onTextPropertyRestrictionNode1(stack["k"], stack["v"])
|
|
}
|
|
|
|
func (c *current) onPhraseNode1(v any) (any, error) {
|
|
return buildStringNode("", v, c.text, c.pos)
|
|
|
|
}
|
|
|
|
func (p *parser) callonPhraseNode1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onPhraseNode1(stack["v"])
|
|
}
|
|
|
|
func (c *current) onWordNode1(v any) (any, error) {
|
|
return buildStringNode("", v, c.text, c.pos)
|
|
|
|
}
|
|
|
|
func (p *parser) callonWordNode1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onWordNode1(stack["v"])
|
|
}
|
|
|
|
func (c *current) onBooleanOperatorNode1() (any, error) {
|
|
return buildOperatorNode(c.text, c.pos)
|
|
|
|
}
|
|
|
|
func (p *parser) callonBooleanOperatorNode1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onBooleanOperatorNode1()
|
|
}
|
|
|
|
func (c *current) onColonOperator1() (any, error) {
|
|
return c.text, nil
|
|
|
|
}
|
|
|
|
func (p *parser) callonColonOperator1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onColonOperator1()
|
|
}
|
|
|
|
func (c *current) onEqualOperator1() (any, error) {
|
|
return c.text, nil
|
|
|
|
}
|
|
|
|
func (p *parser) callonEqualOperator1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onEqualOperator1()
|
|
}
|
|
|
|
func (c *current) onChar1() (any, error) {
|
|
return c.text, nil
|
|
|
|
}
|
|
|
|
func (p *parser) callonChar1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onChar1()
|
|
}
|
|
|
|
func (c *current) onString1(v any) (any, error) {
|
|
return v, nil
|
|
|
|
}
|
|
|
|
func (p *parser) callonString1() (any, error) {
|
|
stack := p.vstack[len(p.vstack)-1]
|
|
_ = stack
|
|
return p.cur.onString1(stack["v"])
|
|
}
|
|
|
|
var (
|
|
// errNoRule is returned when the grammar to parse has no rule.
|
|
errNoRule = errors.New("grammar has no rule")
|
|
|
|
// errInvalidEntrypoint is returned when the specified entrypoint rule
|
|
// does not exit.
|
|
errInvalidEntrypoint = errors.New("invalid entrypoint")
|
|
|
|
// errInvalidEncoding is returned when the source is not properly
|
|
// utf8-encoded.
|
|
errInvalidEncoding = errors.New("invalid encoding")
|
|
|
|
// errMaxExprCnt is used to signal that the maximum number of
|
|
// expressions have been parsed.
|
|
errMaxExprCnt = errors.New("max number of expresssions parsed")
|
|
)
|
|
|
|
// Option is a function that can set an option on the parser. It returns
|
|
// the previous setting as an Option.
|
|
type Option func(*parser) Option
|
|
|
|
// MaxExpressions creates an Option to stop parsing after the provided
|
|
// number of expressions have been parsed, if the value is 0 then the parser will
|
|
// parse for as many steps as needed (possibly an infinite number).
|
|
//
|
|
// The default for maxExprCnt is 0.
|
|
func MaxExpressions(maxExprCnt uint64) Option {
|
|
return func(p *parser) Option {
|
|
oldMaxExprCnt := p.maxExprCnt
|
|
p.maxExprCnt = maxExprCnt
|
|
return MaxExpressions(oldMaxExprCnt)
|
|
}
|
|
}
|
|
|
|
// Entrypoint creates an Option to set the rule name to use as entrypoint.
|
|
// The rule name must have been specified in the -alternate-entrypoints
|
|
// if generating the parser with the -optimize-grammar flag, otherwise
|
|
// it may have been optimized out. Passing an empty string sets the
|
|
// entrypoint to the first rule in the grammar.
|
|
//
|
|
// The default is to start parsing at the first rule in the grammar.
|
|
func Entrypoint(ruleName string) Option {
|
|
return func(p *parser) Option {
|
|
oldEntrypoint := p.entrypoint
|
|
p.entrypoint = ruleName
|
|
if ruleName == "" {
|
|
p.entrypoint = g.rules[0].name
|
|
}
|
|
return Entrypoint(oldEntrypoint)
|
|
}
|
|
}
|
|
|
|
// Statistics adds a user provided Stats struct to the parser to allow
|
|
// the user to process the results after the parsing has finished.
|
|
// Also the key for the "no match" counter is set.
|
|
//
|
|
// Example usage:
|
|
//
|
|
// input := "input"
|
|
// stats := Stats{}
|
|
// _, err := Parse("input-file", []byte(input), Statistics(&stats, "no match"))
|
|
// if err != nil {
|
|
// log.Panicln(err)
|
|
// }
|
|
// b, err := json.MarshalIndent(stats.ChoiceAltCnt, "", " ")
|
|
// if err != nil {
|
|
// log.Panicln(err)
|
|
// }
|
|
// fmt.Println(string(b))
|
|
func Statistics(stats *Stats, choiceNoMatch string) Option {
|
|
return func(p *parser) Option {
|
|
oldStats := p.Stats
|
|
p.Stats = stats
|
|
oldChoiceNoMatch := p.choiceNoMatch
|
|
p.choiceNoMatch = choiceNoMatch
|
|
if p.Stats.ChoiceAltCnt == nil {
|
|
p.Stats.ChoiceAltCnt = make(map[string]map[string]int)
|
|
}
|
|
return Statistics(oldStats, oldChoiceNoMatch)
|
|
}
|
|
}
|
|
|
|
// Debug creates an Option to set the debug flag to b. When set to true,
|
|
// debugging information is printed to stdout while parsing.
|
|
//
|
|
// The default is false.
|
|
func Debug(b bool) Option {
|
|
return func(p *parser) Option {
|
|
old := p.debug
|
|
p.debug = b
|
|
return Debug(old)
|
|
}
|
|
}
|
|
|
|
// Memoize creates an Option to set the memoize flag to b. When set to true,
|
|
// the parser will cache all results so each expression is evaluated only
|
|
// once. This guarantees linear parsing time even for pathological cases,
|
|
// at the expense of more memory and slower times for typical cases.
|
|
//
|
|
// The default is false.
|
|
func Memoize(b bool) Option {
|
|
return func(p *parser) Option {
|
|
old := p.memoize
|
|
p.memoize = b
|
|
return Memoize(old)
|
|
}
|
|
}
|
|
|
|
// AllowInvalidUTF8 creates an Option to allow invalid UTF-8 bytes.
|
|
// Every invalid UTF-8 byte is treated as a utf8.RuneError (U+FFFD)
|
|
// by character class matchers and is matched by the any matcher.
|
|
// The returned matched value, c.text and c.offset are NOT affected.
|
|
//
|
|
// The default is false.
|
|
func AllowInvalidUTF8(b bool) Option {
|
|
return func(p *parser) Option {
|
|
old := p.allowInvalidUTF8
|
|
p.allowInvalidUTF8 = b
|
|
return AllowInvalidUTF8(old)
|
|
}
|
|
}
|
|
|
|
// Recover creates an Option to set the recover flag to b. When set to
|
|
// true, this causes the parser to recover from panics and convert it
|
|
// to an error. Setting it to false can be useful while debugging to
|
|
// access the full stack trace.
|
|
//
|
|
// The default is true.
|
|
func Recover(b bool) Option {
|
|
return func(p *parser) Option {
|
|
old := p.recover
|
|
p.recover = b
|
|
return Recover(old)
|
|
}
|
|
}
|
|
|
|
// GlobalStore creates an Option to set a key to a certain value in
|
|
// the globalStore.
|
|
func GlobalStore(key string, value any) Option {
|
|
return func(p *parser) Option {
|
|
old := p.cur.globalStore[key]
|
|
p.cur.globalStore[key] = value
|
|
return GlobalStore(key, old)
|
|
}
|
|
}
|
|
|
|
// InitState creates an Option to set a key to a certain value in
|
|
// the global "state" store.
|
|
func InitState(key string, value any) Option {
|
|
return func(p *parser) Option {
|
|
old := p.cur.state[key]
|
|
p.cur.state[key] = value
|
|
return InitState(key, old)
|
|
}
|
|
}
|
|
|
|
// ParseFile parses the file identified by filename.
|
|
func ParseFile(filename string, opts ...Option) (i any, err error) {
|
|
f, err := os.Open(filename)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
defer func() {
|
|
if closeErr := f.Close(); closeErr != nil {
|
|
err = closeErr
|
|
}
|
|
}()
|
|
return ParseReader(filename, f, opts...)
|
|
}
|
|
|
|
// ParseReader parses the data from r using filename as information in the
|
|
// error messages.
|
|
func ParseReader(filename string, r io.Reader, opts ...Option) (any, error) {
|
|
b, err := io.ReadAll(r)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
|
|
return Parse(filename, b, opts...)
|
|
}
|
|
|
|
// Parse parses the data from b using filename as information in the
|
|
// error messages.
|
|
func Parse(filename string, b []byte, opts ...Option) (any, error) {
|
|
return newParser(filename, b, opts...).parse(g)
|
|
}
|
|
|
|
// position records a position in the text.
|
|
type position struct {
|
|
line, col, offset int
|
|
}
|
|
|
|
func (p position) String() string {
|
|
return strconv.Itoa(p.line) + ":" + strconv.Itoa(p.col) + " [" + strconv.Itoa(p.offset) + "]"
|
|
}
|
|
|
|
// savepoint stores all state required to go back to this point in the
|
|
// parser.
|
|
type savepoint struct {
|
|
position
|
|
rn rune
|
|
w int
|
|
}
|
|
|
|
type current struct {
|
|
pos position // start position of the match
|
|
text []byte // raw text of the match
|
|
|
|
// state is a store for arbitrary key,value pairs that the user wants to be
|
|
// tied to the backtracking of the parser.
|
|
// This is always rolled back if a parsing rule fails.
|
|
state storeDict
|
|
|
|
// globalStore is a general store for the user to store arbitrary key-value
|
|
// pairs that they need to manage and that they do not want tied to the
|
|
// backtracking of the parser. This is only modified by the user and never
|
|
// rolled back by the parser. It is always up to the user to keep this in a
|
|
// consistent state.
|
|
globalStore storeDict
|
|
}
|
|
|
|
type storeDict map[string]any
|
|
|
|
// the AST types...
|
|
|
|
type grammar struct {
|
|
pos position
|
|
rules []*rule
|
|
}
|
|
|
|
type rule struct {
|
|
pos position
|
|
name string
|
|
displayName string
|
|
expr any
|
|
}
|
|
|
|
type choiceExpr struct {
|
|
pos position
|
|
alternatives []any
|
|
}
|
|
|
|
type actionExpr struct {
|
|
pos position
|
|
expr any
|
|
run func(*parser) (any, error)
|
|
}
|
|
|
|
type recoveryExpr struct {
|
|
pos position
|
|
expr any
|
|
recoverExpr any
|
|
failureLabel []string
|
|
}
|
|
|
|
type seqExpr struct {
|
|
pos position
|
|
exprs []any
|
|
}
|
|
|
|
type throwExpr struct {
|
|
pos position
|
|
label string
|
|
}
|
|
|
|
type labeledExpr struct {
|
|
pos position
|
|
label string
|
|
expr any
|
|
}
|
|
|
|
type expr struct {
|
|
pos position
|
|
expr any
|
|
}
|
|
|
|
type (
|
|
andExpr expr
|
|
notExpr expr
|
|
zeroOrOneExpr expr
|
|
zeroOrMoreExpr expr
|
|
oneOrMoreExpr expr
|
|
)
|
|
|
|
type ruleRefExpr struct {
|
|
pos position
|
|
name string
|
|
}
|
|
|
|
type stateCodeExpr struct {
|
|
pos position
|
|
run func(*parser) error
|
|
}
|
|
|
|
type andCodeExpr struct {
|
|
pos position
|
|
run func(*parser) (bool, error)
|
|
}
|
|
|
|
type notCodeExpr struct {
|
|
pos position
|
|
run func(*parser) (bool, error)
|
|
}
|
|
|
|
type litMatcher struct {
|
|
pos position
|
|
val string
|
|
ignoreCase bool
|
|
want string
|
|
}
|
|
|
|
type charClassMatcher struct {
|
|
pos position
|
|
val string
|
|
basicLatinChars [128]bool
|
|
chars []rune
|
|
ranges []rune
|
|
classes []*unicode.RangeTable
|
|
ignoreCase bool
|
|
inverted bool
|
|
}
|
|
|
|
type anyMatcher position
|
|
|
|
// errList cumulates the errors found by the parser.
|
|
type errList []error
|
|
|
|
func (e *errList) add(err error) {
|
|
*e = append(*e, err)
|
|
}
|
|
|
|
func (e errList) err() error {
|
|
if len(e) == 0 {
|
|
return nil
|
|
}
|
|
e.dedupe()
|
|
return e
|
|
}
|
|
|
|
func (e *errList) dedupe() {
|
|
var cleaned []error
|
|
set := make(map[string]bool)
|
|
for _, err := range *e {
|
|
if msg := err.Error(); !set[msg] {
|
|
set[msg] = true
|
|
cleaned = append(cleaned, err)
|
|
}
|
|
}
|
|
*e = cleaned
|
|
}
|
|
|
|
func (e errList) Error() string {
|
|
switch len(e) {
|
|
case 0:
|
|
return ""
|
|
case 1:
|
|
return e[0].Error()
|
|
default:
|
|
var buf bytes.Buffer
|
|
|
|
for i, err := range e {
|
|
if i > 0 {
|
|
buf.WriteRune('\n')
|
|
}
|
|
buf.WriteString(err.Error())
|
|
}
|
|
return buf.String()
|
|
}
|
|
}
|
|
|
|
// parserError wraps an error with a prefix indicating the rule in which
|
|
// the error occurred. The original error is stored in the Inner field.
|
|
type parserError struct {
|
|
Inner error
|
|
pos position
|
|
prefix string
|
|
expected []string
|
|
}
|
|
|
|
// Error returns the error message.
|
|
func (p *parserError) Error() string {
|
|
return p.prefix + ": " + p.Inner.Error()
|
|
}
|
|
|
|
// newParser creates a parser with the specified input source and options.
|
|
func newParser(filename string, b []byte, opts ...Option) *parser {
|
|
stats := Stats{
|
|
ChoiceAltCnt: make(map[string]map[string]int),
|
|
}
|
|
|
|
p := &parser{
|
|
filename: filename,
|
|
errs: new(errList),
|
|
data: b,
|
|
pt: savepoint{position: position{line: 1}},
|
|
recover: true,
|
|
cur: current{
|
|
state: make(storeDict),
|
|
globalStore: make(storeDict),
|
|
},
|
|
maxFailPos: position{col: 1, line: 1},
|
|
maxFailExpected: make([]string, 0, 20),
|
|
Stats: &stats,
|
|
// start rule is rule [0] unless an alternate entrypoint is specified
|
|
entrypoint: g.rules[0].name,
|
|
}
|
|
p.setOptions(opts)
|
|
|
|
if p.maxExprCnt == 0 {
|
|
p.maxExprCnt = math.MaxUint64
|
|
}
|
|
|
|
return p
|
|
}
|
|
|
|
// setOptions applies the options to the parser.
|
|
func (p *parser) setOptions(opts []Option) {
|
|
for _, opt := range opts {
|
|
opt(p)
|
|
}
|
|
}
|
|
|
|
type resultTuple struct {
|
|
v any
|
|
b bool
|
|
end savepoint
|
|
}
|
|
|
|
const choiceNoMatch = -1
|
|
|
|
// Stats stores some statistics, gathered during parsing
|
|
type Stats struct {
|
|
// ExprCnt counts the number of expressions processed during parsing
|
|
// This value is compared to the maximum number of expressions allowed
|
|
// (set by the MaxExpressions option).
|
|
ExprCnt uint64
|
|
|
|
// ChoiceAltCnt is used to count for each ordered choice expression,
|
|
// which alternative is used how may times.
|
|
// These numbers allow to optimize the order of the ordered choice expression
|
|
// to increase the performance of the parser
|
|
//
|
|
// The outer key of ChoiceAltCnt is composed of the name of the rule as well
|
|
// as the line and the column of the ordered choice.
|
|
// The inner key of ChoiceAltCnt is the number (one-based) of the matching alternative.
|
|
// For each alternative the number of matches are counted. If an ordered choice does not
|
|
// match, a special counter is incremented. The name of this counter is set with
|
|
// the parser option Statistics.
|
|
// For an alternative to be included in ChoiceAltCnt, it has to match at least once.
|
|
ChoiceAltCnt map[string]map[string]int
|
|
}
|
|
|
|
type parser struct {
|
|
filename string
|
|
pt savepoint
|
|
cur current
|
|
|
|
data []byte
|
|
errs *errList
|
|
|
|
depth int
|
|
recover bool
|
|
debug bool
|
|
|
|
memoize bool
|
|
// memoization table for the packrat algorithm:
|
|
// map[offset in source] map[expression or rule] {value, match}
|
|
memo map[int]map[any]resultTuple
|
|
|
|
// rules table, maps the rule identifier to the rule node
|
|
rules map[string]*rule
|
|
// variables stack, map of label to value
|
|
vstack []map[string]any
|
|
// rule stack, allows identification of the current rule in errors
|
|
rstack []*rule
|
|
|
|
// parse fail
|
|
maxFailPos position
|
|
maxFailExpected []string
|
|
maxFailInvertExpected bool
|
|
|
|
// max number of expressions to be parsed
|
|
maxExprCnt uint64
|
|
// entrypoint for the parser
|
|
entrypoint string
|
|
|
|
allowInvalidUTF8 bool
|
|
|
|
*Stats
|
|
|
|
choiceNoMatch string
|
|
// recovery expression stack, keeps track of the currently available recovery expression, these are traversed in reverse
|
|
recoveryStack []map[string]any
|
|
}
|
|
|
|
// push a variable set on the vstack.
|
|
func (p *parser) pushV() {
|
|
if cap(p.vstack) == len(p.vstack) {
|
|
// create new empty slot in the stack
|
|
p.vstack = append(p.vstack, nil)
|
|
} else {
|
|
// slice to 1 more
|
|
p.vstack = p.vstack[:len(p.vstack)+1]
|
|
}
|
|
|
|
// get the last args set
|
|
m := p.vstack[len(p.vstack)-1]
|
|
if m != nil && len(m) == 0 {
|
|
// empty map, all good
|
|
return
|
|
}
|
|
|
|
m = make(map[string]any)
|
|
p.vstack[len(p.vstack)-1] = m
|
|
}
|
|
|
|
// pop a variable set from the vstack.
|
|
func (p *parser) popV() {
|
|
// if the map is not empty, clear it
|
|
m := p.vstack[len(p.vstack)-1]
|
|
if len(m) > 0 {
|
|
// GC that map
|
|
p.vstack[len(p.vstack)-1] = nil
|
|
}
|
|
p.vstack = p.vstack[:len(p.vstack)-1]
|
|
}
|
|
|
|
// push a recovery expression with its labels to the recoveryStack
|
|
func (p *parser) pushRecovery(labels []string, expr any) {
|
|
if cap(p.recoveryStack) == len(p.recoveryStack) {
|
|
// create new empty slot in the stack
|
|
p.recoveryStack = append(p.recoveryStack, nil)
|
|
} else {
|
|
// slice to 1 more
|
|
p.recoveryStack = p.recoveryStack[:len(p.recoveryStack)+1]
|
|
}
|
|
|
|
m := make(map[string]any, len(labels))
|
|
for _, fl := range labels {
|
|
m[fl] = expr
|
|
}
|
|
p.recoveryStack[len(p.recoveryStack)-1] = m
|
|
}
|
|
|
|
// pop a recovery expression from the recoveryStack
|
|
func (p *parser) popRecovery() {
|
|
// GC that map
|
|
p.recoveryStack[len(p.recoveryStack)-1] = nil
|
|
|
|
p.recoveryStack = p.recoveryStack[:len(p.recoveryStack)-1]
|
|
}
|
|
|
|
func (p *parser) print(prefix, s string) string {
|
|
if !p.debug {
|
|
return s
|
|
}
|
|
|
|
fmt.Printf("%s %d:%d:%d: %s [%#U]\n",
|
|
prefix, p.pt.line, p.pt.col, p.pt.offset, s, p.pt.rn)
|
|
return s
|
|
}
|
|
|
|
func (p *parser) in(s string) string {
|
|
p.depth++
|
|
return p.print(strings.Repeat(" ", p.depth)+">", s)
|
|
}
|
|
|
|
func (p *parser) out(s string) string {
|
|
p.depth--
|
|
return p.print(strings.Repeat(" ", p.depth)+"<", s)
|
|
}
|
|
|
|
func (p *parser) addErr(err error) {
|
|
p.addErrAt(err, p.pt.position, []string{})
|
|
}
|
|
|
|
func (p *parser) addErrAt(err error, pos position, expected []string) {
|
|
var buf bytes.Buffer
|
|
if p.filename != "" {
|
|
buf.WriteString(p.filename)
|
|
}
|
|
if buf.Len() > 0 {
|
|
buf.WriteString(":")
|
|
}
|
|
buf.WriteString(fmt.Sprintf("%d:%d (%d)", pos.line, pos.col, pos.offset))
|
|
if len(p.rstack) > 0 {
|
|
if buf.Len() > 0 {
|
|
buf.WriteString(": ")
|
|
}
|
|
rule := p.rstack[len(p.rstack)-1]
|
|
if rule.displayName != "" {
|
|
buf.WriteString("rule " + rule.displayName)
|
|
} else {
|
|
buf.WriteString("rule " + rule.name)
|
|
}
|
|
}
|
|
pe := &parserError{Inner: err, pos: pos, prefix: buf.String(), expected: expected}
|
|
p.errs.add(pe)
|
|
}
|
|
|
|
func (p *parser) failAt(fail bool, pos position, want string) {
|
|
// process fail if parsing fails and not inverted or parsing succeeds and invert is set
|
|
if fail == p.maxFailInvertExpected {
|
|
if pos.offset < p.maxFailPos.offset {
|
|
return
|
|
}
|
|
|
|
if pos.offset > p.maxFailPos.offset {
|
|
p.maxFailPos = pos
|
|
p.maxFailExpected = p.maxFailExpected[:0]
|
|
}
|
|
|
|
if p.maxFailInvertExpected {
|
|
want = "!" + want
|
|
}
|
|
p.maxFailExpected = append(p.maxFailExpected, want)
|
|
}
|
|
}
|
|
|
|
// read advances the parser to the next rune.
|
|
func (p *parser) read() {
|
|
p.pt.offset += p.pt.w
|
|
rn, n := utf8.DecodeRune(p.data[p.pt.offset:])
|
|
p.pt.rn = rn
|
|
p.pt.w = n
|
|
p.pt.col++
|
|
if rn == '\n' {
|
|
p.pt.line++
|
|
p.pt.col = 0
|
|
}
|
|
|
|
if rn == utf8.RuneError && n == 1 { // see utf8.DecodeRune
|
|
if !p.allowInvalidUTF8 {
|
|
p.addErr(errInvalidEncoding)
|
|
}
|
|
}
|
|
}
|
|
|
|
// restore parser position to the savepoint pt.
|
|
func (p *parser) restore(pt savepoint) {
|
|
if p.debug {
|
|
defer p.out(p.in("restore"))
|
|
}
|
|
if pt.offset == p.pt.offset {
|
|
return
|
|
}
|
|
p.pt = pt
|
|
}
|
|
|
|
// Cloner is implemented by any value that has a Clone method, which returns a
|
|
// copy of the value. This is mainly used for types which are not passed by
|
|
// value (e.g map, slice, chan) or structs that contain such types.
|
|
//
|
|
// This is used in conjunction with the global state feature to create proper
|
|
// copies of the state to allow the parser to properly restore the state in
|
|
// the case of backtracking.
|
|
type Cloner interface {
|
|
Clone() any
|
|
}
|
|
|
|
var statePool = &sync.Pool{
|
|
New: func() any { return make(storeDict) },
|
|
}
|
|
|
|
func (sd storeDict) Discard() {
|
|
for k := range sd {
|
|
delete(sd, k)
|
|
}
|
|
statePool.Put(sd)
|
|
}
|
|
|
|
// clone and return parser current state.
|
|
func (p *parser) cloneState() storeDict {
|
|
if p.debug {
|
|
defer p.out(p.in("cloneState"))
|
|
}
|
|
|
|
state := statePool.Get().(storeDict)
|
|
for k, v := range p.cur.state {
|
|
if c, ok := v.(Cloner); ok {
|
|
state[k] = c.Clone()
|
|
} else {
|
|
state[k] = v
|
|
}
|
|
}
|
|
return state
|
|
}
|
|
|
|
// restore parser current state to the state storeDict.
|
|
// every restoreState should applied only one time for every cloned state
|
|
func (p *parser) restoreState(state storeDict) {
|
|
if p.debug {
|
|
defer p.out(p.in("restoreState"))
|
|
}
|
|
p.cur.state.Discard()
|
|
p.cur.state = state
|
|
}
|
|
|
|
// get the slice of bytes from the savepoint start to the current position.
|
|
func (p *parser) sliceFrom(start savepoint) []byte {
|
|
return p.data[start.position.offset:p.pt.position.offset]
|
|
}
|
|
|
|
func (p *parser) getMemoized(node any) (resultTuple, bool) {
|
|
if len(p.memo) == 0 {
|
|
return resultTuple{}, false
|
|
}
|
|
m := p.memo[p.pt.offset]
|
|
if len(m) == 0 {
|
|
return resultTuple{}, false
|
|
}
|
|
res, ok := m[node]
|
|
return res, ok
|
|
}
|
|
|
|
func (p *parser) setMemoized(pt savepoint, node any, tuple resultTuple) {
|
|
if p.memo == nil {
|
|
p.memo = make(map[int]map[any]resultTuple)
|
|
}
|
|
m := p.memo[pt.offset]
|
|
if m == nil {
|
|
m = make(map[any]resultTuple)
|
|
p.memo[pt.offset] = m
|
|
}
|
|
m[node] = tuple
|
|
}
|
|
|
|
func (p *parser) buildRulesTable(g *grammar) {
|
|
p.rules = make(map[string]*rule, len(g.rules))
|
|
for _, r := range g.rules {
|
|
p.rules[r.name] = r
|
|
}
|
|
}
|
|
|
|
func (p *parser) parse(g *grammar) (val any, err error) {
|
|
if len(g.rules) == 0 {
|
|
p.addErr(errNoRule)
|
|
return nil, p.errs.err()
|
|
}
|
|
|
|
// TODO : not super critical but this could be generated
|
|
p.buildRulesTable(g)
|
|
|
|
if p.recover {
|
|
// panic can be used in action code to stop parsing immediately
|
|
// and return the panic as an error.
|
|
defer func() {
|
|
if e := recover(); e != nil {
|
|
if p.debug {
|
|
defer p.out(p.in("panic handler"))
|
|
}
|
|
val = nil
|
|
switch e := e.(type) {
|
|
case error:
|
|
p.addErr(e)
|
|
default:
|
|
p.addErr(fmt.Errorf("%v", e))
|
|
}
|
|
err = p.errs.err()
|
|
}
|
|
}()
|
|
}
|
|
|
|
startRule, ok := p.rules[p.entrypoint]
|
|
if !ok {
|
|
p.addErr(errInvalidEntrypoint)
|
|
return nil, p.errs.err()
|
|
}
|
|
|
|
p.read() // advance to first rune
|
|
val, ok = p.parseRule(startRule)
|
|
if !ok {
|
|
if len(*p.errs) == 0 {
|
|
// If parsing fails, but no errors have been recorded, the expected values
|
|
// for the farthest parser position are returned as error.
|
|
maxFailExpectedMap := make(map[string]struct{}, len(p.maxFailExpected))
|
|
for _, v := range p.maxFailExpected {
|
|
maxFailExpectedMap[v] = struct{}{}
|
|
}
|
|
expected := make([]string, 0, len(maxFailExpectedMap))
|
|
eof := false
|
|
if _, ok := maxFailExpectedMap["!."]; ok {
|
|
delete(maxFailExpectedMap, "!.")
|
|
eof = true
|
|
}
|
|
for k := range maxFailExpectedMap {
|
|
expected = append(expected, k)
|
|
}
|
|
sort.Strings(expected)
|
|
if eof {
|
|
expected = append(expected, "EOF")
|
|
}
|
|
p.addErrAt(errors.New("no match found, expected: "+listJoin(expected, ", ", "or")), p.maxFailPos, expected)
|
|
}
|
|
|
|
return nil, p.errs.err()
|
|
}
|
|
return val, p.errs.err()
|
|
}
|
|
|
|
func listJoin(list []string, sep string, lastSep string) string {
|
|
switch len(list) {
|
|
case 0:
|
|
return ""
|
|
case 1:
|
|
return list[0]
|
|
default:
|
|
return strings.Join(list[:len(list)-1], sep) + " " + lastSep + " " + list[len(list)-1]
|
|
}
|
|
}
|
|
|
|
func (p *parser) parseRule(rule *rule) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseRule " + rule.name))
|
|
}
|
|
|
|
if p.memoize {
|
|
res, ok := p.getMemoized(rule)
|
|
if ok {
|
|
p.restore(res.end)
|
|
return res.v, res.b
|
|
}
|
|
}
|
|
|
|
start := p.pt
|
|
p.rstack = append(p.rstack, rule)
|
|
p.pushV()
|
|
val, ok := p.parseExpr(rule.expr)
|
|
p.popV()
|
|
p.rstack = p.rstack[:len(p.rstack)-1]
|
|
if ok && p.debug {
|
|
p.print(strings.Repeat(" ", p.depth)+"MATCH", string(p.sliceFrom(start)))
|
|
}
|
|
|
|
if p.memoize {
|
|
p.setMemoized(start, rule, resultTuple{val, ok, p.pt})
|
|
}
|
|
return val, ok
|
|
}
|
|
|
|
func (p *parser) parseExpr(expr any) (any, bool) {
|
|
var pt savepoint
|
|
|
|
if p.memoize {
|
|
res, ok := p.getMemoized(expr)
|
|
if ok {
|
|
p.restore(res.end)
|
|
return res.v, res.b
|
|
}
|
|
pt = p.pt
|
|
}
|
|
|
|
p.ExprCnt++
|
|
if p.ExprCnt > p.maxExprCnt {
|
|
panic(errMaxExprCnt)
|
|
}
|
|
|
|
var val any
|
|
var ok bool
|
|
switch expr := expr.(type) {
|
|
case *actionExpr:
|
|
val, ok = p.parseActionExpr(expr)
|
|
case *andCodeExpr:
|
|
val, ok = p.parseAndCodeExpr(expr)
|
|
case *andExpr:
|
|
val, ok = p.parseAndExpr(expr)
|
|
case *anyMatcher:
|
|
val, ok = p.parseAnyMatcher(expr)
|
|
case *charClassMatcher:
|
|
val, ok = p.parseCharClassMatcher(expr)
|
|
case *choiceExpr:
|
|
val, ok = p.parseChoiceExpr(expr)
|
|
case *labeledExpr:
|
|
val, ok = p.parseLabeledExpr(expr)
|
|
case *litMatcher:
|
|
val, ok = p.parseLitMatcher(expr)
|
|
case *notCodeExpr:
|
|
val, ok = p.parseNotCodeExpr(expr)
|
|
case *notExpr:
|
|
val, ok = p.parseNotExpr(expr)
|
|
case *oneOrMoreExpr:
|
|
val, ok = p.parseOneOrMoreExpr(expr)
|
|
case *recoveryExpr:
|
|
val, ok = p.parseRecoveryExpr(expr)
|
|
case *ruleRefExpr:
|
|
val, ok = p.parseRuleRefExpr(expr)
|
|
case *seqExpr:
|
|
val, ok = p.parseSeqExpr(expr)
|
|
case *stateCodeExpr:
|
|
val, ok = p.parseStateCodeExpr(expr)
|
|
case *throwExpr:
|
|
val, ok = p.parseThrowExpr(expr)
|
|
case *zeroOrMoreExpr:
|
|
val, ok = p.parseZeroOrMoreExpr(expr)
|
|
case *zeroOrOneExpr:
|
|
val, ok = p.parseZeroOrOneExpr(expr)
|
|
default:
|
|
panic(fmt.Sprintf("unknown expression type %T", expr))
|
|
}
|
|
if p.memoize {
|
|
p.setMemoized(pt, expr, resultTuple{val, ok, p.pt})
|
|
}
|
|
return val, ok
|
|
}
|
|
|
|
func (p *parser) parseActionExpr(act *actionExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseActionExpr"))
|
|
}
|
|
|
|
start := p.pt
|
|
val, ok := p.parseExpr(act.expr)
|
|
if ok {
|
|
p.cur.pos = start.position
|
|
p.cur.text = p.sliceFrom(start)
|
|
state := p.cloneState()
|
|
actVal, err := act.run(p)
|
|
if err != nil {
|
|
p.addErrAt(err, start.position, []string{})
|
|
}
|
|
p.restoreState(state)
|
|
|
|
val = actVal
|
|
}
|
|
if ok && p.debug {
|
|
p.print(strings.Repeat(" ", p.depth)+"MATCH", string(p.sliceFrom(start)))
|
|
}
|
|
return val, ok
|
|
}
|
|
|
|
func (p *parser) parseAndCodeExpr(and *andCodeExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseAndCodeExpr"))
|
|
}
|
|
|
|
state := p.cloneState()
|
|
|
|
ok, err := and.run(p)
|
|
if err != nil {
|
|
p.addErr(err)
|
|
}
|
|
p.restoreState(state)
|
|
|
|
return nil, ok
|
|
}
|
|
|
|
func (p *parser) parseAndExpr(and *andExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseAndExpr"))
|
|
}
|
|
|
|
pt := p.pt
|
|
state := p.cloneState()
|
|
p.pushV()
|
|
_, ok := p.parseExpr(and.expr)
|
|
p.popV()
|
|
p.restoreState(state)
|
|
p.restore(pt)
|
|
|
|
return nil, ok
|
|
}
|
|
|
|
func (p *parser) parseAnyMatcher(any *anyMatcher) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseAnyMatcher"))
|
|
}
|
|
|
|
if p.pt.rn == utf8.RuneError && p.pt.w == 0 {
|
|
// EOF - see utf8.DecodeRune
|
|
p.failAt(false, p.pt.position, ".")
|
|
return nil, false
|
|
}
|
|
start := p.pt
|
|
p.read()
|
|
p.failAt(true, start.position, ".")
|
|
return p.sliceFrom(start), true
|
|
}
|
|
|
|
func (p *parser) parseCharClassMatcher(chr *charClassMatcher) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseCharClassMatcher"))
|
|
}
|
|
|
|
cur := p.pt.rn
|
|
start := p.pt
|
|
|
|
// can't match EOF
|
|
if cur == utf8.RuneError && p.pt.w == 0 { // see utf8.DecodeRune
|
|
p.failAt(false, start.position, chr.val)
|
|
return nil, false
|
|
}
|
|
|
|
if chr.ignoreCase {
|
|
cur = unicode.ToLower(cur)
|
|
}
|
|
|
|
// try to match in the list of available chars
|
|
for _, rn := range chr.chars {
|
|
if rn == cur {
|
|
if chr.inverted {
|
|
p.failAt(false, start.position, chr.val)
|
|
return nil, false
|
|
}
|
|
p.read()
|
|
p.failAt(true, start.position, chr.val)
|
|
return p.sliceFrom(start), true
|
|
}
|
|
}
|
|
|
|
// try to match in the list of ranges
|
|
for i := 0; i < len(chr.ranges); i += 2 {
|
|
if cur >= chr.ranges[i] && cur <= chr.ranges[i+1] {
|
|
if chr.inverted {
|
|
p.failAt(false, start.position, chr.val)
|
|
return nil, false
|
|
}
|
|
p.read()
|
|
p.failAt(true, start.position, chr.val)
|
|
return p.sliceFrom(start), true
|
|
}
|
|
}
|
|
|
|
// try to match in the list of Unicode classes
|
|
for _, cl := range chr.classes {
|
|
if unicode.Is(cl, cur) {
|
|
if chr.inverted {
|
|
p.failAt(false, start.position, chr.val)
|
|
return nil, false
|
|
}
|
|
p.read()
|
|
p.failAt(true, start.position, chr.val)
|
|
return p.sliceFrom(start), true
|
|
}
|
|
}
|
|
|
|
if chr.inverted {
|
|
p.read()
|
|
p.failAt(true, start.position, chr.val)
|
|
return p.sliceFrom(start), true
|
|
}
|
|
p.failAt(false, start.position, chr.val)
|
|
return nil, false
|
|
}
|
|
|
|
func (p *parser) incChoiceAltCnt(ch *choiceExpr, altI int) {
|
|
choiceIdent := fmt.Sprintf("%s %d:%d", p.rstack[len(p.rstack)-1].name, ch.pos.line, ch.pos.col)
|
|
m := p.ChoiceAltCnt[choiceIdent]
|
|
if m == nil {
|
|
m = make(map[string]int)
|
|
p.ChoiceAltCnt[choiceIdent] = m
|
|
}
|
|
// We increment altI by 1, so the keys do not start at 0
|
|
alt := strconv.Itoa(altI + 1)
|
|
if altI == choiceNoMatch {
|
|
alt = p.choiceNoMatch
|
|
}
|
|
m[alt]++
|
|
}
|
|
|
|
func (p *parser) parseChoiceExpr(ch *choiceExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseChoiceExpr"))
|
|
}
|
|
|
|
for altI, alt := range ch.alternatives {
|
|
// dummy assignment to prevent compile error if optimized
|
|
_ = altI
|
|
|
|
state := p.cloneState()
|
|
|
|
p.pushV()
|
|
val, ok := p.parseExpr(alt)
|
|
p.popV()
|
|
if ok {
|
|
p.incChoiceAltCnt(ch, altI)
|
|
return val, ok
|
|
}
|
|
p.restoreState(state)
|
|
}
|
|
p.incChoiceAltCnt(ch, choiceNoMatch)
|
|
return nil, false
|
|
}
|
|
|
|
func (p *parser) parseLabeledExpr(lab *labeledExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseLabeledExpr"))
|
|
}
|
|
|
|
p.pushV()
|
|
val, ok := p.parseExpr(lab.expr)
|
|
p.popV()
|
|
if ok && lab.label != "" {
|
|
m := p.vstack[len(p.vstack)-1]
|
|
m[lab.label] = val
|
|
}
|
|
return val, ok
|
|
}
|
|
|
|
func (p *parser) parseLitMatcher(lit *litMatcher) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseLitMatcher"))
|
|
}
|
|
|
|
start := p.pt
|
|
for _, want := range lit.val {
|
|
cur := p.pt.rn
|
|
if lit.ignoreCase {
|
|
cur = unicode.ToLower(cur)
|
|
}
|
|
if cur != want {
|
|
p.failAt(false, start.position, lit.want)
|
|
p.restore(start)
|
|
return nil, false
|
|
}
|
|
p.read()
|
|
}
|
|
p.failAt(true, start.position, lit.want)
|
|
return p.sliceFrom(start), true
|
|
}
|
|
|
|
func (p *parser) parseNotCodeExpr(not *notCodeExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseNotCodeExpr"))
|
|
}
|
|
|
|
state := p.cloneState()
|
|
|
|
ok, err := not.run(p)
|
|
if err != nil {
|
|
p.addErr(err)
|
|
}
|
|
p.restoreState(state)
|
|
|
|
return nil, !ok
|
|
}
|
|
|
|
func (p *parser) parseNotExpr(not *notExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseNotExpr"))
|
|
}
|
|
|
|
pt := p.pt
|
|
state := p.cloneState()
|
|
p.pushV()
|
|
p.maxFailInvertExpected = !p.maxFailInvertExpected
|
|
_, ok := p.parseExpr(not.expr)
|
|
p.maxFailInvertExpected = !p.maxFailInvertExpected
|
|
p.popV()
|
|
p.restoreState(state)
|
|
p.restore(pt)
|
|
|
|
return nil, !ok
|
|
}
|
|
|
|
func (p *parser) parseOneOrMoreExpr(expr *oneOrMoreExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseOneOrMoreExpr"))
|
|
}
|
|
|
|
var vals []any
|
|
|
|
for {
|
|
p.pushV()
|
|
val, ok := p.parseExpr(expr.expr)
|
|
p.popV()
|
|
if !ok {
|
|
if len(vals) == 0 {
|
|
// did not match once, no match
|
|
return nil, false
|
|
}
|
|
return vals, true
|
|
}
|
|
vals = append(vals, val)
|
|
}
|
|
}
|
|
|
|
func (p *parser) parseRecoveryExpr(recover *recoveryExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseRecoveryExpr (" + strings.Join(recover.failureLabel, ",") + ")"))
|
|
}
|
|
|
|
p.pushRecovery(recover.failureLabel, recover.recoverExpr)
|
|
val, ok := p.parseExpr(recover.expr)
|
|
p.popRecovery()
|
|
|
|
return val, ok
|
|
}
|
|
|
|
func (p *parser) parseRuleRefExpr(ref *ruleRefExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseRuleRefExpr " + ref.name))
|
|
}
|
|
|
|
if ref.name == "" {
|
|
panic(fmt.Sprintf("%s: invalid rule: missing name", ref.pos))
|
|
}
|
|
|
|
rule := p.rules[ref.name]
|
|
if rule == nil {
|
|
p.addErr(fmt.Errorf("undefined rule: %s", ref.name))
|
|
return nil, false
|
|
}
|
|
return p.parseRule(rule)
|
|
}
|
|
|
|
func (p *parser) parseSeqExpr(seq *seqExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseSeqExpr"))
|
|
}
|
|
|
|
vals := make([]any, 0, len(seq.exprs))
|
|
|
|
pt := p.pt
|
|
state := p.cloneState()
|
|
for _, expr := range seq.exprs {
|
|
val, ok := p.parseExpr(expr)
|
|
if !ok {
|
|
p.restoreState(state)
|
|
p.restore(pt)
|
|
return nil, false
|
|
}
|
|
vals = append(vals, val)
|
|
}
|
|
return vals, true
|
|
}
|
|
|
|
func (p *parser) parseStateCodeExpr(state *stateCodeExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseStateCodeExpr"))
|
|
}
|
|
|
|
err := state.run(p)
|
|
if err != nil {
|
|
p.addErr(err)
|
|
}
|
|
return nil, true
|
|
}
|
|
|
|
func (p *parser) parseThrowExpr(expr *throwExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseThrowExpr"))
|
|
}
|
|
|
|
for i := len(p.recoveryStack) - 1; i >= 0; i-- {
|
|
if recoverExpr, ok := p.recoveryStack[i][expr.label]; ok {
|
|
if val, ok := p.parseExpr(recoverExpr); ok {
|
|
return val, ok
|
|
}
|
|
}
|
|
}
|
|
|
|
return nil, false
|
|
}
|
|
|
|
func (p *parser) parseZeroOrMoreExpr(expr *zeroOrMoreExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseZeroOrMoreExpr"))
|
|
}
|
|
|
|
var vals []any
|
|
|
|
for {
|
|
p.pushV()
|
|
val, ok := p.parseExpr(expr.expr)
|
|
p.popV()
|
|
if !ok {
|
|
return vals, true
|
|
}
|
|
vals = append(vals, val)
|
|
}
|
|
}
|
|
|
|
func (p *parser) parseZeroOrOneExpr(expr *zeroOrOneExpr) (any, bool) {
|
|
if p.debug {
|
|
defer p.out(p.in("parseZeroOrOneExpr"))
|
|
}
|
|
|
|
p.pushV()
|
|
val, _ := p.parseExpr(expr.expr)
|
|
p.popV()
|
|
// whether it matched or not, consider it a match
|
|
return val, true
|
|
}
|